Inadequate fixation can often lead to complications involving nonunion or malunion. Careful determination of the type fixation should be evaluated during preoperative evaluation. In a study by Hutchinson, he has shown that the ideal mechanical construct would be an anterior plate and with a lag screw (Fig. 29.7), which is similar to the biomechanical data for first MPJ arthrodesis [34]. It is important to note that lateral plate with compression screw was similar in stiffness. While this construct is sufficient for open procedures, it may not be relevant for arthroscopic procedures, which requires percutaneous fixation. It has been shown that when performing arthroscopic ankle arthrodesis in vivo, the fastest rate of union was achieved with three parallel screws placed medially from the distal tibia into the talus [35]. While this construct has shown to lead to a faster time to fusion, it has not been shown to be biomechanically more stable. In patients with a suspected infectious process, it may be advantageous to utilize external fixation, as internal fixation often will propagate the infectious process (Fig. 29.8).

While the ankle arthrodesis is quite versatile with minimal contraindications, it is essential that as the surgeon, we critically analyze the patient to determine whether or not the procedure would be best for them. As stated above it is essential to perform not only a thorough history and physical but also a social and socioeconomic evaluation of the patient prior to surgery. Often complications can occur because we are focused more on the patient’s radiographic findings and deformity rather than on the patient as a whole. It is the recommendation of this author that patients with multiple comorbidities be critically evaluated to determine whether or not they are candidates for elective ankle arthrodesis; it cannot be understated that a thorough history and physical examination should be performed prior to surgery. Laboratory testing to determine if an infection is present should be performed especially if the patient has sustained an open fracture or any type of trauma that is unknown, or history of surgery. All revision should be worked up for an infectious process. Circulatory, hematopoietic , metabolic, and nutritional profiles should be established and corrected to decrease healing deficits prior to any surgery.

Once the diagnosis of a nonunion has been made, it is vital to identify the cause. A root cause analysis may be performed to determine the etiology of the nonunion. Achieving this can prevent future surgical errors as well as prevent a reoccurrence nonunion. Identification of the type of nonunion is also helpful in creating a treatment plan. Treatment should also be based upon the patient’s symptoms. Often a nonunion may not be painful; in this case, surgery is not indicated. In these cases the patient should be educated as to potential hardware failure, or other complications that may occur in the future. If the nonunion is asymptomatic and stable and the patient is neuropathic, a rigid brace should be prescribed to maintain correction [37].

The surgical plan should begin with the incisional approach. A thorough inspection of the soft tissue envelope should be performed as well as a review of vasculature in the potential surgical site. Often it is necessary to place the incision through the previous surgical site when percutaneous hardware removal cannot be achieved. If percutaneous hardware removal can be achieved, it may advantageous to plan an incision that will allow adequate access to the nonunion and through non-traumatized skin. Incision placement, when possible, should respect the angiosomes of the lower extremity. Anterior incision placement should be avoided if possible, as it may compromise this angiosome and is the reason for a 25% reported dehiscence rate in ankle arthroplasty [38].

Lateral or posterior incisional approaches tend to be safe and reliable. Hanson reported no complications in his series of ten patients undergoing pantalar arthrodesis through a posterior approach (Fig. 29.10).

Once the surgical approach has been determined, attention should be directed to the osseous component of the surgery. A plethora of fixation, biologic, and grafting options are available to the surgeon in preparation for surgery. There is no general consensus on which combination of options is most advantageous. The treatment plant should be patient specific and therefore may result in a variety of combinations of fixation, biologic, and nonbiologic augmentations as well as bone grafting. Typically if the nonunion is a simple hypertrophic type, without any other suspected pathology other than poor technique and/or fixation , the nonunion may be treated with takedown and aggressive resection followed by either bone grafting or biologic augmentation, with rigid internal fixation. If the patient has multiple comorbidities that may have contributed to the nonunion, it is recommended that the underlying malady be addressed. This might include tighter control of a patient’s Hba1c, smoking cessation, weight loss, or discontinuation of DMARDS. Only once these issues are addressed, it is recommended to proceed with the surgical plan.

There are over 40 documented fixation techniques described in the literature which fall into two major categories: internal fixation which is the most common and includes crossing screws, plate fixation, intramedullary nails, and combinations of these [39]. The other is external fixation, which is typically reserved for complex planar deformities, Charcot, and in patients who have suspected or known infection. When using screw fixation it has been advocated that a three-screw tripod technique be utilized. This technique is commonly employed for primary fusions; however, it is the opinion of the author that given the biomechanical data comparing screw with plate and screw fixation, the later technique be employed to ensure the most biomechanically stable construct. Patients who are identified with a nonunion may often have concomitant subtalar joint pain with degeneration of the joint. Use of intramedullary retrograde nail or precontoured anatomic tibiotalocalcaneal (TTC) locking plates should be considered in this patient population. A recent biomechanical evaluation of intramedullary nail versus locking TTC fusion plate with TTC augmentation screw in cadaveric models showed higher final rigidity in the locking plate group [40]. While plate fixation may be more biomechanically stable, it often requires a large exposure; therefore, it may not be optimal if there is concern for the soft tissue envelope. If possible a posterior approach with a posterior locking TTC plate would be an ideal approach if plate fixation is the choice of fixation. In cases where minimal exposure is desired, intramedullary nail fixation may be utilized successfully. In a retrospective review of ten cases of nonunion, a minimally invasive retrograde locking nail gave excellent results with a 90% union rate [41]. The use of a blade plate has also been advocated by many authors for complex ankle or TTC fusions with reported successful union rates as high as 90% [42]. Biomechanically this construct has been shown to be inferior to crossing screw constructs in a sawbone model [43] (Fig. 29.11).

External fixation can be a valuable tool for treatment of nonunions especially when infection is suspected. External fixation provides rigid immobilization with resistance to multiplane stresses especially torsional and axial load while allowing access to soft tissue defects and early weight bearing (Fig. 29.12). If there is significant bone loss after resection, external fixation may but utilize with proximal corticotomy and distraction osteogenesis to lengthen and then compression distally at the nonunion site. External fixation is the author’s fixation of choice in treatment of septic nonunions. External fixation may be considered for aseptic nonunions as well; however, in a retrospective review of 26 nonunions treated with external fixation, Kitaoka reported a satisfactory result with only 77% of the nonunions going on to successful fusion and fair or poor results in nearly half the patients [44].

The past decade has seen significant advances in the field of orthobiologics. Bone biology may be manipulated to promote healing potential through osteoconductive or osteoinductive mediators. Osteoconductive products provide a scaffold for ingrowth of bone (Fig. 29.13). More recently research has been focused on osteoinductive products that stimulate osteoprogenitor stem cells to differentiate into osteoblasts. Bone morphogenetic protein (BMP) has been the most widely studied type of the osteoinductive cell mediators. Bibbo showed a 96% union rate in 108 high-risk fusion patients using recombinant human BMP-2 (rhBMP-2) [45]. Equally impressive, Fourman reported a 93% union rate in his rhBMP-2 group versus a 53% union rate in patients without the growth factor in complex ankle fusions [46]. Allogenic stem cells have also been reported in the literature as a safe and effective biologic augmentation, although it has not performed as well as autogenous grafting [47].